Liz Stokes, Natasha Simons, Tom Honeyman (Australian Research Data Commons), Chris Erdmann (Library Carpentry/The Carpentries/California Digital Library), Sharyn Wise (University of Technology, Sydney), Josh Quan, Scott Peterson, Amy Neeser (UC Berkeley)
To translate FAIR principles into useable concepts for research-facing support staff (e.g. librarians).
- Library staff who provide research support
- Those who want to know more about FAIR and how it could be applied to libraries
- Translating FAIR speak to library speak (What is it? Why do I need to know? What do I tell researchers?)
- Identifying ways to improve the ‘FAIRness’ of your library
- Understanding that FAIR data helps us be better stewards of our own resources
Thing 1: Why should librarians care about FAIR?
There’s a lot of hype about the FAIR Data Principles. But why should librarians care? For starters, libraries have a strong tradition in describing resources, providing access and building collections, and providing support for the long-term stewardship of digital resources. Building on their specific knowledge and expertise, librarians should feel confident with making research data FAIR. So how can you and your library get started with the FAIR principles?
- Read LIBER’s Implementing FAIR Principles: the role of Libraries at https://libereurope.eu/wp-content/uploads/2017/12/LIBER-FAIR-Data.pdf (5 minute read)
- Where is your library at in regard to the section on ‘getting started with FAIR’?
- Where are you at in your own understanding of the FAIR Data Principles?
Read TIB’s expanded walk through of the FAIR data principles (18 minute read). Take note of the roles for scientists and repositories and think about where you might use these examples in discussion with researchers and colleagues.
Thing 2: How FAIR are your data?
The FAIR Principles are easily understood in theory but more challenging when applied in practice. In this exercise, you will be using the Australian Research Data Commons (ARDC) Data self-assessment tool to assess the ‘FAIRness’ of one of your library’s datasets.
- Select a metadata record from your library’s collection (e.g. your institutional repository) that describes a published dataset. If your library or institution doesn’t have a repository for research data, choose one from re3data.org
- Open the ARDC FAIR Data Assessment tool and run your chosen dataset against the tool to assess its ‘FAIRness’.
- How FAIR was your chosen dataset?
- How easy was it to apply the FAIR criteria to your dataset?
- What things need to happen in order to improve the ‘FAIRness’ of your chosen dataset?
Try your hand at other tools like the CSIRO 5 star data rating tool and the DANS FAIR data assessment tool.
Thing 3: Do you teach FAIR to your researchers?
How FAIR aware are your researchers? Does your library incorporate FAIR into researcher training?
- Go to existing data management/data sharing training you provide to Graduates, Higher Degree Researchers (HDRs) or other researchers. For example, review the Duke Graduate School’s Responsible Conduct of Research topics page.
- Review how well the 15 FAIR Principles are covered in this training and adjust accordingly.
Thing 4: Is FAIR built into library practice and policy?
Your library may do a great job advocating the FAIR Data Principles to researchers but how well have the Principles been incorporated into library practice and policy?
- Review your library or institutional policies regarding research data management and digital preservation with the FAIR Principles in mind. Consider that in most cases library policy will have been written before the advent of FAIR. Are revisions required?
- Review the data repository managed by your library or institution. How well does it support FAIR Data?
- Review your library’s Data Management Planning tool. Does it have features that support the FAIR Data Principles or are changes required?
Thing 5: Are your library staff trained in FAIR?
Reusing the wide range of openly available training materials available in the FAIR Data Principles e.g. you could start here.
- Conduct a skills and knowledge audit regarding FAIR with your library team.
- Based on the audit, identify gaps in FAIR skills and knowledge.
- Design a training program that can fill the identified gaps. To help build your program, read the blog post, A Carpentries based approach to teaching FAIR data and software principles.
- Reusing the wide range of openly available training materials available in the FAIR Data Principles e.g. you could start here.
- Plugging into librarian networks to find opportunities to collaborate e.g. #datalibs on Twitter, Data Curation Network.
Thing 6: Are digital libraries FAIR?
While the FAIR Principles are designed for data, considering their potential application in a broader context is useful. For example, think about what criteria might be applied to assess the ‘FAIRness’ of digital libraries. Considerations might include:
- Persistent identifiers
- Open access vs. paid access
- Provenance information / metadata
- Author credibility
- Versioning information
- License / reuse information
- Usage statistics (number of times downloaded)
- Select one of these digital libraries (or another of your choice):
- Search/browse the catalogue of items.
- Does the library display reuse permissions/licenses on how to use the item?
- Is there provenance information?
- Are persistent identifiers used?
A number of FAIR principles make reference to “metadata”. What is metadata, how is it relevant to FAIR and does your library support the kind of metadata specified in the FAIR Data Principles?
- Watch this video in which the Metadata Librarian explains metadata (3 mins)
- Select three metadata records at random for datasets held in your library or repository collection.
- Open the checklist produced for use at the EUDAT summer school and see if you can check off those that reference metadata against the records you selected.
- Make a list of what metadata elements could be improved in your library records to enable better support for FAIR.
Thing 8: Does your library support FAIR identifiers?
The FAIR data principles call for open, standardised protocols for accessing data via a persistent identifier. Persistent identifiers are crucial for the findability and identification of research, researchers and for tracking impact metrics. So how well does your library support persistent identifiers?
- Find out how well your library supports ORCIDs and DOIs:
- Do your library systems support the identification of researchers via an ORCID? Do you authenticate against the ORCID registry? Do you have an ORCID?
- Do your library systems, such as your institutional repository, support the issuing of Digital Object Identifiers (DOIs) for research data and related materials?
- What other types of persistent identifiers do you think your library should support? Why or why not?
If your library supports the minting of DOIs for research data and related materials, is there more that you could do in this regard? Check out A Data Citation Roadmap for Scholarly Repositories and determine how much of the roadmap you can check off your list and how much is yet to do.
Thing 9: Does your library support FAIR protocols?
For (meta)data to be accessible it should ideally be available via a standard protocol. Think of protocols in terms of borrowing a book: there are a number of expectations that the library lays out in order to proceed. You have to identify yourself using a library card, you have to bring the book to the checkout desk, and in return you walk out of the library with a demagnetised book and receipt reminding you when you have to return the book by. Accessing the books in the library means that you must learn and abide by the rules for accessing books.
- Familiarise yourself with APIs by completing Thing 19 of the ANDS 23 (research data) Things
- Consider the APIs your library provides to enable access (meta)data for data and related materials. Are they up to scratch or are improvements required?
Thing 10: Next steps for your library in supporting FAIR
In Thing 1 you read LIBER’s Implementing FAIR Principles: the role of Libraries. You considered what your library needed to do in order to better support FAIR data. In Thing 10 we will create a list of outstanding action items.
- Write a list of what your library is currently doing to support and promote the FAIR Data Principles.
- Now compare this to the list in the LIBER document. Where are the gaps and what can you do to fill these?
- Create an action plan to improve FAIR support at your library!
- Incorporating all that you learnt and the progress that you made in “doing” this Top 10 FAIR Things!
- Reaching out to #datalibs on Twitter for tips and support for your action plan